In contrast to host RNAs, which, if they can be spliced, are exported from the nucleus only after splicing, retroviral RNAs must be exported in both spliced and unspliced forms. Lentiviruses accomplish this by encoding a protein, Rev, which interacts with a highly structured RNA element, the Rev response element (RRE), within the Env coding region. This interaction leads to the nuclear export of RRE-containing RNAs. The RRE is composed of several discrete domains; it contains a high-affinity or primary Rev binding site, called site IIB, and a second preferred binding site on a very long, imperfect stem (domain I). Several crystallographic structures of Rev have been determined, either free or bound to small fragments of the RRE. _____Despite the importance of this problem, Rev-RRE interaction is still not well understood; one challenging question is how Rev specifically interacts with RRE-containing RNAs, since these RNAs are in a minority in the nucleus of the infected cell and since Rev can bind RNA rather nonspecifically. As the Rev-RRE interaction is essential for HIV-1 replication, it is an attractive target for antiviral therapy. In a fruitful collaboration with a structural biologist, Dr. Yun-Xing Wang (Structural Biophysics Laboratory, NCI), we have recently made significant progress on this problem. Using small-angle X-ray scattering (SAXS), Dr. Wang showed that the RRE assumes an A shape, with the primary Rev-binding sites 55 angstroms apart on opposite legs of the A. This is the width of a Rev dimer; thus, this fit could explain Rev specificity for the RRE. We developed a quantitative assay for RRE function and determined the functional contributions of individual RRE domains. The data support this essential role for the A shape, and also suggest that the entire structure is under strain; some domains function to stabilize it. We identified four base pairs necessary for function and are analyzing the reason for this. One widely discussed RRE model, the jellyfish model, seems incompatible with our data. Our assay can also be used to test candidate inhibitors of RRE function. _____Another poorly understood area in contemporary retrovirus biology is the function of the gammaretroviral protein, glycogag (ggag). Except for an 88-residue stretch at its N-terminus, the sequence of ggag is identical to that of Gag. However, there is no evidence suggesting that its function is related to that of Gag or is dependent upon the Gag portion of the sequence. In fact, ggag appears to be a gammaretroviral accessory protein, despite the widely held belief that gammaretroviruses are simple retroviruses lacking accessory proteins. It was accepted for many years that ggag is not important for MLV replication in cell culture, but is critical for successful virus replication in mice. Suggested functions of ggag include improving the fidelity of virus particle assembly; enhancing the stability of the mature viral capsid; directing virus production to lipid rafts in virus-producing cells; protecting MLV from mA3; and enhancing the pathogenicity of MLV. Recent reports show that it can complement a Nef defect in HIV-1 and that, like Nef, it counteracts the effects of the cellular protein Serinc5. We find that ggag profoundly enhances the specific infectivity of MLV with some retroviral Env proteins, protecting these viruses from the effects of Serinc5. Conversely, it is detrimental for MLV carrying Ebola virus glycoprotein, while Serinc5 promotes infectivity of these particles. These positive and negative effects of ggag on viral infectivity are exerted at the cell entry step. Thus, our studies on a gammaretroviral accessory protein have led to surprising and potentially important new findings on Ebola virus entry. We are investigating the effects of ggag and Serinc5 on cellular and viral lipid composition. _____Corresponds to Rein Project 4 in the July 2016 site visit report of the HIV Dynamics and Replication Program]